Enhancing Video Transformers for Action Understanding with VLM-aided Training

Open paper
Benchmark Model Rank Results
action-classification-on-kinetics-400FTP-UniFormerV2-L/14–Acc@1: 93.4
action-recognition-in-videos-on-ucf101FTP-UniFormerV2-L/14–3-fold Accuracy: 99.7

Rank counts only results with a code link.